TY - JOUR T1 - <strong>Automatic Lesion Detection and Segmentation of <sup>18</sup></strong><strong>FET PET in gliomas : A Full 3D U-Net Convolutional Neural Network Study.</strong> JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 330 LP - 330 VL - 59 IS - supplement 1 AU - Paul Blanc-Durand AU - Axel Van Der Gucht AU - Niklaus Schaefer AU - Emmanuel Itti AU - John Prior Y1 - 2018/05/01 UR - http://jnm.snmjournals.org/content/59/supplement_1/330.abstract N2 - 330Introduction: Amino-acids positron emission tomography (PET) is increasingly used in the diagnostic workup of patients with gliomas, including differential diagnosis, evaluation of tumor extension, treatment planning and follow-up. Recently, the progress of computer vision and machine learning has been translated for medical imaging. Aim was to demonstrate the feasibility of an automated 18F-FET PET lesion detection and segmentation relying on a full 3-D U-Net Convolutional Neural Network (CNN). Methods: All dynamic 18F-FET PET brain image volumes were temporally realigned to the first dynamic acquisition, coregistered and spatially normalized onto the Montreal Neurological Institute template. Ground truth segmentations were obtained using manual delineation and thresholding (1.3 × background). The volumetric convolution neural network was implemented based on a modified Keras implementation of a U-net library with 3 layers for the encoding and decoding paths. Dice similarity coefficient (DSC) was used as an accuracy measure of segmentation. Results: 37 patients were included (26 (71%) in the training set and 11 (29%) in the validation set). All 11 lesions were accurately detected with no false positive, resulting in a sensitivity and a specificity for the detection at the tumoral level of 100%. After 150 epochs DSC reached 0.7911 in the validation set and a DSC of 0.7924 in the training set. After morphological dilatation and fixed thresholding of the predicted U-net mask a substantial improvement of the DSC to 0.8231 (+ 4.1%) [JoP1] [PB2] was noted. At the voxel level, this segmentation led to a 0.88 sensitivity [95% CI, 87.1 to, 88.2%] a 0.99 specificity [99.9 to 99.9%],,a 0.78 positive predictive value: [76.9 to 78.3%], and a 0.99 negative predictive value [99.9 to 99.9%]. Conclusions: With relatively high performance, we propose the first full 3-D automated procedure for segmentation of 18F-FET PET brain images of patients with different gliomas using a U-Net CNN architecture. ER -